Energy efficiency analysis method, device, equipment and storage medium for photovoltaic power station

Abstract

The present invention relates to the technical field of energy efficiency optimization, and discloses an energy efficiency analysis method, device, equipment and storage medium of a photovoltaic power station, including the following steps: S101, collecting power station characteristic data of the photovoltaic power station, and representing it through a first characteristic sequence; S102, processing the first characteristic sequence to obtain a second characteristic sequence; S103, constructing an output power prediction model according to the second characteristic sequence, and the output of the output power prediction model is the output power of the photovoltaic power station in a second preset time period; S104, generating optimal inductance and capacitance values ​​according to the output power prediction model combined with an improved crown porcupine optimization algorithm. The present invention maximizes the efficiency of light energy utilization while ensuring voltage and current stability through an improved GRU model and an improved crown porcupine optimization algorithm, and has the significant advantages of adapting to complex environments, improving system reliability and grid compatibility.

Description

Technical Field

[0001] The present invention belongs to the technical field of energy efficiency optimization, and specifically relates to an energy efficiency analysis method, device, equipment and storage medium for a photovoltaic power station. Background Art

[0002] As a clean and renewable energy system, photovoltaic power stations have been widely used around the world in recent years. However, the energy efficiency and current and voltage stability of photovoltaic power stations are key factors affecting their performance and economic benefits. Energy efficiency refers to the efficiency of photovoltaic power stations in converting solar energy into electrical energy, while current and voltage stability refers to the ability of the system to maintain stable output under various operating conditions.

[0003] In current technologies, the energy efficiency optimization of photovoltaic power stations mainly focuses on the following aspects: optimizing the working point of photovoltaic modules through the maximum power point tracking (MPPT) algorithm to improve the utilization efficiency of the modules; improving the conversion efficiency of DC to AC through efficient inverter design and control algorithms; and real-time monitoring of system operating parameters for operating status evaluation. However, these technologies still have obvious defects in practical applications. First, most optimization methods only target a single link and fail to achieve coordinated optimization of the overall performance of the system; second, the fixed parameter configuration method is difficult to adapt to complex environmental changes, such as drastic fluctuations in light intensity or module temperature; in addition, the existing technology does not pay enough attention to the stability of current and voltage, making it difficult to balance the relationship between efficiency and stability, which easily leads to fluctuations beyond the safe range. Summary of the invention

[0004] The present invention provides an energy efficiency analysis method, device, equipment and storage medium for a photovoltaic power station, which solves the technical problems in related technologies that they focus too much on the optimization of a single link, lack global coordination capabilities, and are difficult to dynamically adapt to complex environmental changes.

[0005] The present invention provides a method for analyzing the energy efficiency of a photovoltaic power station, comprising the following steps:

[0006] Step S101, collecting power station characteristic data of the photovoltaic power station within a first preset time period T at a preset time interval t, represented by a first characteristic sequence, the power station characteristic data includes: photovoltaic module voltage, photovoltaic module current, photovoltaic power station voltage, photovoltaic power station current, inductance value, capacitance value, ambient temperature, photovoltaic module temperature, light intensity and output power, wherein T and t are custom parameters;

[0007] Step S102, processing the first feature sequence to obtain a second feature sequence;

[0008] Step S103, constructing an output power prediction model according to the second characteristic sequence, wherein the output of the output power prediction model is the output power of the photovoltaic power station within a second preset time period;

[0009] Step S104, generating an optimal parameter combination based on the output power prediction model combined with the improved crown porcupine optimization algorithm, wherein the optimal parameter combination represents the optimal inductance and capacitance values ​​obtained by maximizing the light energy efficiency and the stability of the photovoltaic power station voltage and photovoltaic power station current within the second preset time period.

[0010] Further, the first feature sequence is processed to obtain a second feature sequence, and the specific steps include:

[0011] Step S201, using linear interpolation to fill in missing values ​​in the first feature sequence;

[0012] Step S202, use 3 The principle identifies outliers in the first feature sequence and replaces them with the median of all non-outliers in the first feature sequence;

[0013] Step S203, use The values ​​of the first feature sequence are processed, where represents the standard deviation of the Gaussian filter, a represents the time point, represents the Gaussian function value at the ath time point, and exp represents the exponential function;

[0014] Step S204: construct an output power change rate feature according to the first feature sequence, and splice it after the processed first feature sequence, wherein the calculation formula of the output power change rate feature is: ,in, Indicates the output power change rate characteristics, and denote the output power in the first characteristic sequence at the ath and a-1th time points, respectively. and represents the time of the ath and a-1th time points;

[0015] Step S205, using a z-score normalization method to normalize the concatenated first feature sequence to obtain a second feature sequence.

[0016] Furthermore, the output power prediction model is constructed based on the improved GRU, and the structure of the output power prediction model includes: an input layer, an improved GRU layer, a Dropout layer and an output layer;

[0017] The input of the input layer is the second feature sequence;

[0018] The improved GRU layer includes N hidden units, each hidden unit inputs the nth sequence unit of the second feature sequence, and outputs the hidden state at the current moment. Each hidden unit contains a set of improved gating mechanisms, and the improved gating mechanisms include: update gate, reset gate, enhancement gate, candidate hidden state and hidden state, where N=T / t, 1≤n≤N;

[0019] The enhancement gate is used to control the degree of fusion between the previous hidden state and the current input;

[0020] The Dropout layer is located after the improved GRU layer and is used to randomly discard part of the hidden state data;

[0021] The output layer includes a classifier, and the classification space of the classifier is the output power of the photovoltaic power station within the second preset time period.

[0022] Furthermore, the calculation formula of the improved GRU layer includes:

[0023] ;

[0024] ;

[0025] ;

[0026] ;

[0027] ;

[0028] in, Represents the nth sequence unit of the second feature sequence currently input, and denote the hidden states of the nth and n-1th hidden unit outputs, respectively. , , and Represent the update gate, reset gate, enhancement gate and candidate hidden state respectively, , and They represent the first weight parameter, the second weight parameter and the bias parameter corresponding to the update gate respectively, , and They represent the first weight parameter, the second weight parameter and the bias parameter corresponding to the reset gate respectively, , and They represent the first weight parameter, the second weight parameter and the bias parameter corresponding to the enhancement gate respectively, , and They represent the first weight parameter, the second weight parameter and the bias parameter corresponding to the candidate hidden state respectively, sigmoid represents the sigmoid activation function, Represents point-wise multiplication.

[0029] Furthermore, the Dropout layer is inserted between the hidden units, that is, between the output of the n-1th hidden unit and the input of the nth hidden unit. The calculation formula of the Dropout layer includes:

[0030] ;

[0031] ;

[0032] in, represents the data of the part of the hidden state discarded by the nth hidden unit, express It follows the Bernoulli distribution. A binary mask indicating whether to retain the hidden state. A value of 1 indicates retention, a value of 0 indicates discarding, and p indicates the probability of retention.

[0033] Furthermore, the optimal parameter combination is generated according to the output power prediction model combined with the improved crested porcupine optimization algorithm. The specific steps include:

[0034] Step S301, initializing a population satisfying the constraint conditions based on Levy flight, the population consisting of Q individuals;

[0035] Each individual is identified by the first eigenvector Indicates, where L represents the inductance value and C represents the capacitance value;

[0036] The constraints include:

[0037] The inductance value cannot exceed the allowable range of the photovoltaic module;

[0038] The capacitance value cannot exceed the allowable range of the photovoltaic module;

[0039] Step S302, constructing an objective function by combining an output power prediction model;

[0040] Step S303, calculate the fitness value of each individual in the population according to the objective function, and the calculation formula of the objective function is:

[0041] ;

[0042] in, represents the fitness value of the objective function, represents the output power obtained by the output power prediction model, and Indicates the PV module voltage and PV module current, and Indicates the voltage and current of the photovoltaic power station. and Indicates the maximum voltage and current that a photovoltaic module can output under standard test conditions. Standard test conditions indicate that the light intensity is 1000 , the ambient temperature is 25℃, and denote the first temperature coefficient and the second temperature coefficient respectively, Indicates the ambient temperature, represents the ambient temperature under standard test conditions, G represents the light intensity, Indicates the light intensity under standard test conditions;

[0043] Step S304, generating a random number ranging from 0 to 1, and when the random number is less than a first preset threshold, performing global exploration, otherwise performing local development, the first preset threshold being a custom parameter;

[0044] Step S305, after the current iteration is completed, the number of populations is reduced according to the cyclic population reduction technique, and the first preset threshold is updated;

[0045] Step S306, when the current number of iterations is greater than or equal to the maximum number of iterations, the individual with the largest fitness in the initialized population is used as the optimal parameter combination, otherwise, return to step S303, where the maximum number of iterations is a custom parameter.

[0046] Furthermore, a population satisfying the constraints is initialized based on the Levy flight. The specific steps include:

[0047] The range of the first eigenvector of the individual is obtained according to the constraint condition, and the mathematical expression of the range of the first eigenvector is: , and Indicates the minimum and maximum inductance values, and Indicates the minimum and maximum values ​​of capacitance, Indicates that P is in the vector and vector within the scope;

[0048] Step S3011, randomly generating an initial population according to uniform distribution;

[0049] Step S3012: for each individual's first eigenvector, a step vector is generated using Levy distribution, wherein the step vector is calculated as follows: , represents the step vector, represents the step size scaling factor, u and v represent random values ​​from a normal distribution, represents the absolute value of v, represents the first exponential coefficient;

[0050] Step S3013, updating the first eigenvector of each individual according to the step vector of each individual, and the calculation formula for updating the first eigenvector is: , represents the updated first eigenvector;

[0051] Step S3014, perform boundary check on each individual, and if the individual is out of range, adjust it back to the boundary value according to the adjustment formula, the adjustment formula is: , , where max represents the maximum value operation and min represents the minimum value operation.

[0052] The present invention provides an energy efficiency analysis device for a photovoltaic power station, comprising:

[0053] A data acquisition unit is used to collect power station characteristic data of the photovoltaic power station within a first preset time period T according to a preset time interval t, and the power station characteristic data is represented by a first characteristic sequence, and the power station characteristic data includes: photovoltaic module voltage, photovoltaic module current, photovoltaic power station voltage, photovoltaic power station current, inductance value, capacitance value, ambient temperature, photovoltaic module temperature, light intensity, output power, wherein T and t are custom parameters;

[0054] A data processing unit, used for processing the first feature sequence to obtain a second feature sequence;

[0055] An output power prediction unit, used to construct an output power prediction model according to the second characteristic sequence, wherein the output of the output power prediction model is the output power of the photovoltaic power station within a second preset time period;

[0056] A parameter combination generating unit is used to generate an optimal parameter combination according to an output power prediction model combined with an improved crown porcupine optimization algorithm, wherein the optimal parameter combination represents the optimal inductance and capacitance values ​​obtained by maximizing the light energy efficiency and maximizing the stability of the photovoltaic power station voltage and the photovoltaic power station current within a second preset time period.

[0057] The present invention provides an electronic device, comprising a processor and a memory storing a computer program, wherein the processor implements the above-mentioned method for analyzing the energy efficiency of a photovoltaic power station when executing the computer program.

[0058] The present invention provides a computer-readable storage medium, in which a program is stored. When the program is run on a computer, the computer executes the above-mentioned method for analyzing the energy efficiency of a photovoltaic power station.

[0059] The beneficial effects of the present invention are as follows: the present invention constructs an output power prediction model based on the improved GRU model, the improved GRU introduces an enhancement gate, which can dynamically adjust the fusion of input features and hidden states, significantly improving the ability to capture complex time series relationships in photovoltaic data, and at the same time, the addition of the Dropout layer effectively prevents overfitting of the model, thereby improving the prediction accuracy and generalization ability;

[0060] The present invention combines the improved crown porcupine optimization algorithm and uses the output power prediction model to generate the optimal inductance and capacitance parameter combination, thereby maximizing the light energy efficiency and ensuring the stability of the output voltage and current of the photovoltaic power station. This collaborative optimization not only improves the power generation efficiency, but also enhances the system's operational reliability and grid compatibility.

[0061] The objective function of the present invention takes into account the changes in the temperature and light intensity of the photovoltaic components during design, so that the optimization results can dynamically adapt to the complex environmental conditions in actual operation. Regardless of light fluctuations or temperature changes, the system can quickly adjust the optimization parameters to maintain high energy efficiency and stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 is a flow chart of the energy efficiency analysis method of a photovoltaic power station of the present invention;

[0063] Figure 2 is a flow chart of processing the first feature sequence of the present invention;

[0064] Figure 3 It is a flow chart of the improved crested porcupine optimization algorithm of the present invention. DETAILED DESCRIPTION

[0065] The subject matter described herein will now be discussed with reference to example embodiments. It should be understood that the discussion of these embodiments is only to enable those skilled in the art to better understand and implement the subject matter described herein, and the functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the contents of this specification. Each example may omit, replace or add various processes or components as needed. In addition, the features described relative to some examples may also be combined in other examples.

[0066] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in one or more embodiments of the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in one or more embodiments of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0067] like Figure 1-Figure 3 As shown, the energy efficiency analysis method of a photovoltaic power station includes the following steps:

[0068] Step S101, collecting power station characteristic data of the photovoltaic power station within a first preset time period T at a preset time interval t, represented by a first characteristic sequence, the power station characteristic data includes: photovoltaic module voltage, photovoltaic module current, photovoltaic power station voltage, photovoltaic power station current, inductance value, capacitance value, ambient temperature, photovoltaic module temperature, light intensity and output power, wherein T and t are custom parameters;

[0069] Step S102, processing the first feature sequence to obtain a second feature sequence;

[0070] Step S103, constructing an output power prediction model according to the second characteristic sequence, wherein the output of the output power prediction model is the output power of the photovoltaic power station within a second preset time period;

[0071] Step S104, generating an optimal parameter combination based on the output power prediction model combined with the improved crown porcupine optimization algorithm, wherein the optimal parameter combination represents the optimal inductance and capacitance values ​​obtained by maximizing the light energy efficiency and the stability of the photovoltaic power station voltage and photovoltaic power station current within the second preset time period.

[0072] In one embodiment of the present invention, the first feature sequence is processed to obtain the second feature sequence, and the specific steps include:

[0073] Step S201, using linear interpolation to fill missing values ​​in the first feature sequence. Specifically, the linear interpolation calculation formula is: ,in, represents the missing values ​​after filling. and Respectively represent the values ​​of the previous and next time points of the missing value. Indicates the time point corresponding to the missing value, and Indicates the time point before and after the missing value;

[0074] Step S202, use 3 The principle is to identify outliers in the first feature sequence and replace them with the median of all non-outliers in the first feature sequence. Specifically, the mean and standard deviation of the first feature sequence are calculated to determine the normal data range: , detect outliers outside the normal data range, and replace the detected outliers with the median to ensure the stability and accuracy of the data. is the mean, is the standard deviation;

[0075] Step S203, use The values ​​of the first feature sequence are processed, where represents the standard deviation of the Gaussian filter, a represents the time point, represents the Gaussian function value at the ath time point, and exp represents the exponential function;

[0076] Step S204: construct an output power change rate feature according to the first feature sequence, and splice it after the processed first feature sequence, wherein the calculation formula of the output power change rate feature is: ,in, Indicates the output power change rate characteristics, and denote the output power in the first characteristic sequence at the ath and a-1th time points, respectively. and represents the time of the ath and a-1th time points;

[0077] Step S205, using a z-score normalization method to normalize the concatenated first feature sequence to obtain a second feature sequence.

[0078] In one embodiment of the present invention, the output power prediction model is constructed based on the improved GRU, and the structure of the output power prediction model includes: an input layer, an improved GRU layer, a Dropout layer and an output layer;

[0079] The input of the input layer is the second feature sequence;

[0080] The improved GRU layer includes N hidden units, each hidden unit inputs the nth sequence unit of the second feature sequence, and outputs the hidden state at the current moment. Each hidden unit contains a set of improved gating mechanisms, and the improved gating mechanisms include: update gate, reset gate, enhancement gate, candidate hidden state and hidden state, where N=T / t, 1≤n≤N;

[0081] The enhancement gate is used to control the degree of fusion between the hidden state of the previous moment and the current input, further improving the model's ability to learn key features;

[0082] The Dropout layer is located after the improved GRU layer and is used to randomly discard some hidden state data to prevent overfitting;

[0083] The output layer includes a classifier, and the classification space of the classifier is the output power of the photovoltaic power station within the second preset time period.

[0084] In one embodiment of the present invention, the calculation formula of the improved GRU layer includes:

[0085] ;

[0086] ;

[0087] ;

[0088] ;

[0089] ;

[0090] in, Represents the nth sequence unit of the second feature sequence currently input, and denote the hidden states of the nth and n-1th hidden unit outputs, respectively. , , and Represent the update gate, reset gate, enhancement gate and candidate hidden state respectively, , and They represent the first weight parameter, the second weight parameter and the bias parameter corresponding to the update gate respectively, , and They represent the first weight parameter, the second weight parameter and the bias parameter corresponding to the reset gate respectively, , and They represent the first weight parameter, the second weight parameter and the bias parameter corresponding to the enhancement gate respectively, , and They represent the first weight parameter, the second weight parameter and the bias parameter corresponding to the candidate hidden state respectively, sigmoid represents the sigmoid activation function, Represents point-wise multiplication.

[0091] The enhancement gate is added to the gating mechanism to further optimize the fusion process of input and hidden state in the GRU structure. The traditional GRU mainly relies on the update gate and reset gate to control the update and memory of the hidden state, while the introduction of the enhancement gate can provide more flexibility and capabilities. Specifically, the enhancement gate can dynamically adjust the degree of dependence on the input features, thereby better capturing the impact of key variables (such as voltage, current and ambient temperature) in photovoltaic data.

[0092] In one embodiment of the present invention, the Dropout layer is inserted between hidden units, that is, between the output of the n-1th hidden unit and the input of the nth hidden unit, and the calculation formula of the Dropout layer includes:

[0093] ;

[0094] ;

[0095] in, represents the data of the part of the hidden state discarded by the nth hidden unit, express It follows the Bernoulli distribution. A binary mask indicating whether to retain the hidden state. A value of 1 indicates retention, a value of 0 indicates discarding, and p indicates the probability of retention.

[0096] In one embodiment of the present invention, a cross entropy loss function is used as the loss function of the output power prediction model, and an adaptive moment estimation optimizer is used to adjust the weight of the output power prediction model. The formula of the adaptive moment estimation optimizer is:

[0097] ;

[0098] in, represents the weight parameter in the output power prediction model, represents the learning rate coefficient, and They represent first-order momentum and second-order momentum respectively. The first-order momentum represents the direction and trend of the gradient, and the second-order momentum represents the oscillation amplitude of the gradient. Represents a random number ranging from 0 to 1.

[0099] In one embodiment of the present invention, the optimal parameter combination is generated according to the output power prediction model combined with the improved crested porcupine optimization algorithm, and the specific steps include:

[0100] Step S301, initializing a population satisfying the constraint conditions based on Levy flight, the population consisting of Q individuals;

[0101] Each individual is identified by the first eigenvector Indicates, where L represents the inductance value and C represents the capacitance value;

[0102] The constraints include:

[0103] The inductance value cannot exceed the allowable range of the photovoltaic module;

[0104] The capacitance value cannot exceed the allowable range of the photovoltaic module;

[0105] Step S302, constructing an objective function by combining an output power prediction model;

[0106] Step S303, calculating the fitness value of each individual in the population according to the objective function;

[0107] Step S304, generating a random number ranging from 0 to 1, and when the random number is less than a first preset threshold, performing global exploration, otherwise performing local development;

[0108] The first preset threshold is a custom variable parameter, which decreases as the number of iterations increases. The formula of the first preset threshold is: ,in represents the first preset threshold when the population iteration number is k+1, represents the first preset threshold when the number of population iterations is k, β represents a decreasing factor between 0 and 1, preferably, the decreasing factor is set to 0.99, and the first preset threshold is set to 0.8;

[0109] Step S3041, during global exploration, a random number ranging from 0 to 1 is generated, and when the random number is less than a second preset threshold, the first defense strategy is adopted, otherwise the second defense strategy is adopted;

[0110] Step S3042: during local development, a random number ranging from 0 to 1 is generated. When the random number is less than the second preset threshold, the third defense strategy is adopted, otherwise the fourth defense strategy is adopted.

[0111] The formula for the first defense strategy is:

[0112] ;

[0113] The formula for the second defense strategy is:

[0114] ;

[0115] The formula for the third defense strategy is:

[0116] ;

[0117] ;

[0118] ;

[0119] ;

[0120] The formula for the fourth defense strategy is:

[0121] ;

[0122] in represents the position of the i-th individual when the number of iterations is k+1, represents the position of the i-th individual when the number of iterations is k, represents random numbers that follow a normal distribution, , , , and rand is a random number between 0 and 1, represents the global optimal solution when the number of iterations is k, represents the position of the predator when the number of iterations is k, r is a random integer between 1 and Q, represents the position of the rth individual when the number of iterations is k, represents a binary vector consisting entirely of 0s, then , the predator no longer moves towards the crested porcupine, when When contains 1, the second defense strategy is based on two random individuals and To determine whether the predator is approaching or moving away from the crested porcupine, r1, r2, and r3 are random integers between 1 and Q. represents the position of the r3th individual when the number of iterations is k, represents the first vector parameter used to control the search direction, represents the defense factor when the number of iterations is k, It represents the odor diffusion factor when the number of iterations is k, where k is the current number of iterations. is the maximum number of iterations, represents the fitness of the i-th individual when the number of iterations is k, As the second parameter, avoid division by 0. is the speed convergence coefficient, exp represents the exponential function, represents the exploration coefficient, represents normal distribution;

[0123] Step S305, after the current iteration is completed, the number of populations is reduced according to the cyclic population reduction technique, and the first preset threshold is updated;

[0124] The formula for the cyclic population reduction technique is: ,in represents the number of individuals in the updated population, Represents the minimum value of the population, preferably, Set to 10, represents the number of individuals in the population when the number of iterations is k, Represents a loop variable, preferably, Set to 2, Represents the remainder operation;

[0125] Step S306, when the current number of iterations is greater than or equal to the maximum number of iterations, the individual with the largest fitness in the initialized population is used as the optimal parameter combination, otherwise return to step S303, where the maximum number of iterations is a custom parameter, preferably, the maximum number of iterations is set to 200.

[0126] In one embodiment of the present invention, a population satisfying constraint conditions is initialized based on Levy flight, and the specific steps include:

[0127] The range of the first eigenvector of the individual is obtained according to the constraint conditions. The mathematical expression of the range of the first eigenvector is: , and Indicates the minimum and maximum inductance values, and Indicates the minimum and maximum values ​​of capacitance, Indicates that P is in the vector and vector within the scope;

[0128] Step S3011, randomly generate an initial population according to uniform distribution, and the calculation formula for generating the initial population is: , ,in , , and represents a uniformly distributed random number, represents uniform distribution;

[0129] Step S3012: for each individual's first eigenvector, a step vector is generated using Levy distribution, wherein the step vector is calculated as follows: , represents the step vector, represents the step size scaling factor, u and v represent random values ​​from a normal distribution, represents the absolute value of v, represents the first exponential coefficient, preferably, Set to 1.5;

[0130] Step S3013, updating the first eigenvector of each individual according to the step vector of each individual, and the calculation formula for updating the first eigenvector is: , represents the updated first eigenvector;

[0131] Step S3014, perform boundary check on each individual, and adjust it back to the boundary value according to the adjustment formula when the individual exceeds the range. For example, when the inductance value of the individual exceeds the maximum inductance value after the last individual is updated, the inductance value of the individual is replaced with the maximum inductance value. The adjustment formula is: , , where max represents the maximum value operation and min represents the minimum value operation.

[0132] In one embodiment of the present invention, the calculation formula of the objective function is:

[0133] ;

[0134] in, represents the fitness value of the objective function, represents the output power obtained by the output power prediction model, and Indicates the PV module voltage and PV module current, and Indicates the voltage and current of the photovoltaic power station. and Indicates the maximum voltage and current that a photovoltaic module can output under standard test conditions. Standard test conditions indicate that the light intensity is 1000 , the ambient temperature is 25℃, and denote the first temperature coefficient and the second temperature coefficient respectively, Indicates the ambient temperature, represents the ambient temperature under standard test conditions, i.e. 25°C, G represents the light intensity, Indicates the light intensity under standard test conditions, i.e. 1000 , and They are used to estimate the expected current and expected voltage of photovoltaic modules respectively. By maximizing the fitness value, the light energy efficiency and the stability of the photovoltaic power station voltage and photovoltaic power station current can be maximized.

[0135] An embodiment of the present invention provides an energy efficiency analysis device for a photovoltaic power station, comprising:

[0136] A data acquisition unit is used to collect power station characteristic data of the photovoltaic power station within a first preset time period T according to a preset time interval t, and the power station characteristic data is represented by a first characteristic sequence, and the power station characteristic data includes: photovoltaic module voltage, photovoltaic module current, photovoltaic power station voltage, photovoltaic power station current, inductance value, capacitance value, ambient temperature, photovoltaic module temperature, light intensity, output power, wherein T and t are custom parameters;

[0137] A data processing unit, used for processing the first feature sequence to obtain a second feature sequence;

[0138] An output power prediction unit, used to construct an output power prediction model according to the second characteristic sequence, wherein the output of the output power prediction model is the output power of the photovoltaic power station within a second preset time period;

[0139] A parameter combination generating unit is used to generate an optimal parameter combination according to an output power prediction model combined with an improved crown porcupine optimization algorithm, wherein the optimal parameter combination represents the optimal inductance and capacitance values ​​obtained by maximizing the light energy efficiency and maximizing the stability of the photovoltaic power station voltage and the photovoltaic power station current within a second preset time period.

[0140] An embodiment of the present invention provides an electronic device, including a processor and a memory storing a computer program. When the processor executes the computer program, the energy efficiency analysis method of the photovoltaic power station provided by the above embodiments is implemented.

[0141] An embodiment of the present invention further provides a computer-readable storage medium, in which a program is stored. When the program is executed on a computer, the computer executes the above-mentioned method for analyzing the energy efficiency of a photovoltaic power station.

[0142] The above describes an embodiment of the present embodiment, but the present embodiment is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present embodiment, ordinary technicians in this field can also make many forms, all of which are protected by the present embodiment.

Claims

1. A method for analyzing energy efficiency of a photovoltaic power station, characterized in that: The following steps are involved: Step S101, collecting power station characteristic data of the photovoltaic power station within a first preset time period T at a preset time interval t, represented by a first characteristic sequence, the power station characteristic data includes: photovoltaic module voltage, photovoltaic module current, photovoltaic power station voltage, photovoltaic power station current, inductance value, capacitance value, ambient temperature, photovoltaic module temperature, light intensity and output power, wherein T and t are custom parameters; Step S102, processing the first feature sequence to obtain a second feature sequence; Step S103, constructing an output power prediction model according to the second characteristic sequence, wherein the output of the output power prediction model is the output power of the photovoltaic power station within a second preset time period; Step S104, generating an optimal parameter combination according to the output power prediction model combined with the improved crested porcupine optimization algorithm, wherein the optimal parameter combination represents the optimal inductance and capacitance values ​​obtained by maximizing the light energy efficiency and maximizing the stability of the photovoltaic power station voltage and the photovoltaic power station current within the second preset time period; Among them, the specific steps of the improved crown porcupine optimization algorithm include: Step S301, initializing a population satisfying the constraint conditions based on Levy flight, the population consisting of Q individuals; Each individual is represented by a first eigenvector P = (L, C), where L represents the inductance value and C represents the capacitance value; Step S302, constructing an objective function by combining an output power prediction model; Step S303, calculate the fitness value of each individual in the population according to the objective function, and the calculation formula of the objective function is: Among them, F represents the fitness value of the objective function, P out represents the output power obtained by the output power prediction model, V DC and I DC Represents the PV module voltage and PV module current, V AC and I AC Represents the voltage and current of the photovoltaic power station, V oc,max and I oc,max Indicates the maximum voltage and current that a photovoltaic module can output under standard test conditions. The standard test conditions indicate that the light intensity is 1000W / m 2 , the ambient temperature is 25℃, β V and β I Represent the first temperature coefficient and the second temperature coefficient, T vir Indicates the ambient temperature, T ref represents the ambient temperature under standard test conditions, G represents the light intensity, G ref Indicates the light intensity under standard test conditions. and They are used to estimate the expected current and expected voltage of the photovoltaic modules respectively. By maximizing the fitness value, the light energy efficiency and the stability of the photovoltaic power station voltage and photovoltaic power station current can be maximized; Step S304, generating a random number ranging from 0 to 1, and when the random number is less than a first preset threshold, performing global exploration, otherwise performing local development, the first preset threshold being a custom parameter; Step S305, after the current iteration is completed, the number of populations is reduced according to the cyclic population reduction technique, and the first preset threshold is updated; Step S306, when the current number of iterations is greater than or equal to the maximum number of iterations, the individual with the largest fitness in the initialized population is used as the optimal parameter combination, otherwise, return to step S303, where the maximum number of iterations is a custom parameter.

2. The method for analyzing energy efficiency of a photovoltaic power station according to claim 1, characterized in that: The first feature sequence is processed to obtain a second feature sequence, and the specific steps include: Step S201, using linear interpolation to fill in missing values ​​in the first feature sequence; Step S202, using the 3σ principle to identify outliers in the first feature sequence, and replacing the outliers with the median of all non-outliers in the first feature sequence; Step S203, use Process the values ​​of the first feature sequence, where σ represents the standard deviation of the Gaussian filter, a represents the time point, y(a,σ) represents the Gaussian function value at the a-th time point, and exp represents the exponential function; Step S204: construct an output power change rate feature according to the first feature sequence, and splice it after the processed first feature sequence, wherein the calculation formula of the output power change rate feature is: Among them, R P Indicates the output power change rate characteristics, P a and P a-1 denote the output power in the first characteristic sequence at the ath and a-1th time points, respectively, t a and t a-1 Indicates the time of the ath and a-1th time points; Step S205, using a z-score normalization method to normalize the concatenated first feature sequence to obtain a second feature sequence.

3. The method for analyzing energy efficiency of a photovoltaic power station according to claim 1, characterized in that: The output power prediction model is constructed based on the improved GRU. The structure of the output power prediction model includes: input layer, improved GRU layer, Dropout layer and output layer; The input of the input layer is the second feature sequence; The improved GRU layer includes N hidden units, each hidden unit inputs the nth sequence unit of the second feature sequence, and outputs the hidden state at the current moment. Each hidden unit contains a set of improved gating mechanisms, and the improved gating mechanisms include: update gate, reset gate, enhancement gate, candidate hidden state and hidden state, where N = T / t, 1≤n≤N; The enhancement gate is used to control the degree of fusion between the previous hidden state and the current input; The Dropout layer is located after the improved GRU layer and is used to randomly discard part of the hidden state data; The output layer includes a classifier, and the classification space of the classifier is the output power of the photovoltaic power station within the second preset time period.

4. The method for analyzing energy efficiency of a photovoltaic power station according to claim 3, characterized in that: The calculation formula of the improved GRU layer includes: Among them, x n represents the nth sequence unit of the second feature sequence currently input, h n and h n-1 Respectively represent the hidden states of the nth and n-1th hidden unit outputs, z n 、r n 、e n and Represent the update gate, reset gate, enhancement gate and candidate hidden state respectively, and They represent the first weight parameter, the second weight parameter and the bias parameter corresponding to the update gate respectively, and They represent the first weight parameter, the second weight parameter and the bias parameter corresponding to the reset gate respectively, and They represent the first weight parameter, the second weight parameter and the bias parameter corresponding to the enhancement gate respectively, and They respectively represent the first weight parameter, the second weight parameter and the bias parameter corresponding to the candidate hidden state, sigmoid represents the sigmoid activation function, and ⊙ represents point-by-point multiplication.

5. The method for analyzing energy efficiency of a photovoltaic power station according to claim 4, characterized in that: The Dropout layer is inserted between hidden units, that is, between the output of the n-1th hidden unit and the input of the nth hidden unit. The calculation formula of the Dropout layer includes: m n ~Bernoulli(p); in, Represents the data of the part of the hidden state discarded by the nth hidden unit, m n ~Bernoilli(p) represents m n Follows Bernoulli distribution, m n A binary mask indicating whether to retain the hidden state. A value of 1 indicates retention, a value of 0 indicates discarding, and p indicates the probability of retention.

6. The method for analyzing energy efficiency of a photovoltaic power station according to claim 1, characterized in that: Initialize a population that meets the constraints based on Levy flight. The specific steps include: According to the constraint condition, the range of the first eigenvector of the individual is obtained. The mathematical expression of the range of the first eigenvector is: P∈[(L min , C min ), (L max , C max )],L min and L max Indicates the minimum and maximum inductance values, C min and C max Represents the minimum and maximum values ​​of capacitance, P∈[(L min , C min ), (L max , C max )] indicates that P is in the vector (L min , C min ) and vector (L max , C max ) Step S3011, randomly generating an initial population according to uniform distribution; Step S3012: for each individual's first eigenvector, a step vector is generated using Levy distribution, wherein the step vector is calculated as follows: ΔP represents the step vector, α represents the step scaling factor, u and v represent random values ​​in the normal distribution, |v| represents the absolute value of v, and λ represents the first exponential coefficient; Step S3013, updating the first eigenvector of each individual according to the step vector of each individual, and the calculation formula for updating the first eigenvector is: represents the updated first eigenvector; Step S3014, perform boundary check on each individual, and adjust it back to the boundary value according to the adjustment formula when the individual exceeds the range. The adjustment formula is: L = min (max (L, L min ), L max ), C = min(max(C, C min ), C max ), where max represents the maximum value operation and min represents the minimum value operation.

7. An energy efficiency analysis device for a photovoltaic power station, used to implement the energy efficiency analysis method for a photovoltaic power station as claimed in any one of claims 1 to 6, characterized in that: include: A data acquisition unit is used to collect power station characteristic data of the photovoltaic power station within a first preset time period T according to a preset time interval t, and the power station characteristic data is represented by a first characteristic sequence, and the power station characteristic data includes: photovoltaic module voltage, photovoltaic module current, photovoltaic power station voltage, photovoltaic power station current, inductance value, capacitance value, ambient temperature, photovoltaic module temperature, light intensity, output power, wherein T and t are custom parameters; A data processing unit, used for processing the first feature sequence to obtain a second feature sequence; An output power prediction unit, used to construct an output power prediction model according to the second characteristic sequence, wherein the output of the output power prediction model is the output power of the photovoltaic power station within a second preset time period; A parameter combination generating unit is used to generate an optimal parameter combination according to an output power prediction model combined with an improved crown porcupine optimization algorithm, wherein the optimal parameter combination represents the optimal inductance and capacitance values ​​obtained by maximizing the light energy efficiency and maximizing the stability of the photovoltaic power station voltage and the photovoltaic power station current within a second preset time period.

8. An electronic device comprising a processor and a memory storing a computer program, characterized in that: When the processor executes the computer program, the energy efficiency analysis method for a photovoltaic power station according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, which, when executed by a processor, implements the energy efficiency analysis method for a photovoltaic power station according to any one of claims 1 to 6.

Images